中文核心期刊
CSCD来源期刊
中国科技核心期刊
RCCSE中国核心学术期刊

重庆交通大学学报(自然科学版) ›› 2012, Vol. 31 ›› Issue (2): 349-352.DOI: 10.3969 /j.issn.1674-0696.2012.02.40

• • 上一篇    下一篇

基于小波网络的船舶柴油机燃油系统故障诊断

印洪浩,彭中波   

  1. 重庆交通大学航海学院,重庆400074
  • 收稿日期:2011-09-09 修回日期:2011-11-07 出版日期:2012-04-15 发布日期:2014-10-31
  • 作者简介:印洪浩( 1974 - ) ,男,重庆人,讲师,博士研究生,主要从事轮机工程方面的教学和研究工作。E-mail: yinhonghao163@163. com。
  • 基金资助:
    重庆市教育委员会自然科学基金项目( KJ00402)

Fault Diagnosis of Marine Diesel Engine Fuel System Based on Wavelet Neural Network

Yin Honghao,Peng Zhongbo   

  1. School of Maritime,Chongqing Jiaotong University,Chongqing 400074,China
  • Received:2011-09-09 Revised:2011-11-07 Online:2012-04-15 Published:2014-10-31

摘要: 采用小波变换和BP 神经网络的辅助式结合,通过小波变换提取故障特征向量作为BP 神经网络的输入值,设 计并组建了小波神经网络; 利用小波变换模极大值分析高压油管燃油压力信号的奇异性,提取故障特征向量; 根据 故障采集数据并建立学习样本,通过网络训练建立BP 神经网络输入和输出间良好的非线性映射,进而通过特征向 量输入BP 神经网络来诊断故障。实验数据分析表明: 该方法具有良好的诊断效果。

关键词: 小波分析, 神经网络, 柴油机, 燃油系统, 故障诊断

Abstract: A wavelet neural network have been designed and built up. The singularity of fuel pressure signal is analyzed by wavelet transform modulus maxima to extract fault feature vectors. According to the data sampled from diesel engine fuel system working conditions,learning samples are obtained. Accordingly,nonlinear mapping between the neural network inputs and outputs have been well established via network training. Afterwards,fault diagnosis is achieved based on the input of feature vectors. According to the analysis and verification from the experimental data,this method is of good diagnosis effect.

Key words: wavelet analysis, neural network, diesel engine, fuel system, fault diagnosis

中图分类号: